A ridgeline plot (also known as Joy Plot, named after the Joy Division album cover) displays the distribution of multiple groups by stacking partially overlapping density curves vertically. This creates a mountain ridge appearance that allows efficient comparison of many distributions simultaneously while maintaining a compact and visually striking presentation.
""" anyplot.ai
ridgeline-basic: Basic Ridgeline Plot
Library: plotnine 0.15.7 | Python 3.13.14
Quality: 89/100 | Updated: 2026-07-25
"""
import os
import numpy as np
import pandas as pd
from plotnine import (
aes,
annotate,
element_blank,
element_line,
element_rect,
element_text,
geom_line,
geom_ribbon,
geom_text,
ggplot,
labs,
scale_fill_gradient,
scale_y_continuous,
theme,
theme_minimal,
)
from scipy import stats
# Theme tokens
THEME = os.getenv("ANYPLOT_THEME", "light")
PAGE_BG = "#FAF8F1" if THEME == "light" else "#1A1A17"
INK = "#1A1A17" if THEME == "light" else "#F0EFE8"
INK_SOFT = "#4A4A44" if THEME == "light" else "#B8B7B0"
ACCENT = "#009E73" # Imprint palette position 1 — focal annotation accent
# Data - Monthly temperature distributions for a temperate climate
np.random.seed(42)
months = ["Jan", "Feb", "Mar", "Apr", "May", "Jun", "Jul", "Aug", "Sep", "Oct", "Nov", "Dec"]
temp_params = {
"Jan": (2, 3),
"Feb": (4, 3),
"Mar": (8, 5),
"Apr": (13, 4),
"May": (18, 4),
"Jun": (22, 3),
"Jul": (25, 3),
"Aug": (24, 3),
"Sep": (20, 4.5),
"Oct": (14, 4),
"Nov": (8, 4),
"Dec": (4, 3),
}
# Generate raw samples for KDE
data = []
for month in months:
mean, std = temp_params[month]
values = np.random.normal(mean, std, 200)
for v in values:
data.append({"month": month, "temp": v})
df = pd.DataFrame(data)
# Compute KDE density curves for ridgeline layout
x_range = np.linspace(-10, 40, 300)
ridge_scale = 2.5
# Trim each month's near-zero-density tails before building ridge_df so the
# ribbon fill (and the top-edge line drawn separately below) only cover the
# visible bump, instead of a sliver extending across the full x_range. The
# threshold mask alone isn't guaranteed contiguous -- gaussian_kde's tail
# estimate can ripple back above a lenient threshold far from the mode -- so
# walk outward from the density peak and stop at the first drop below
# threshold, keeping only that single contiguous run for geom_line to trace.
density_data = []
for i, month in enumerate(months):
month_data = df[df["month"] == month]["temp"]
kde = stats.gaussian_kde(month_data)
density = kde(x_range)
density_scaled = density / density.max() * ridge_scale
threshold = 0.05 * ridge_scale
peak_idx = int(np.argmax(density_scaled))
left = peak_idx
while left > 0 and density_scaled[left - 1] > threshold:
left -= 1
right = peak_idx
while right < len(density_scaled) - 1 and density_scaled[right + 1] > threshold:
right += 1
x_visible = x_range[left : right + 1]
density_visible = density_scaled[left : right + 1]
for x, d in zip(x_visible, density_visible, strict=True):
density_data.append(
{"x": x, "ymin": float(i), "ymax": float(i) + d, "group": month, "month_idx": float(i) / 11.0}
)
ridge_df = pd.DataFrame(density_data)
ridge_df["group"] = pd.Categorical(ridge_df["group"], categories=months, ordered=True)
# Peak label data: placed at July's baseline level (y=jul_idx) to the right
# of where the Jul ridge tapers off — clearly within July's y-band on the axis
jul_idx = months.index("Jul")
peak_df = pd.DataFrame([{"x": 34.5, "y": float(jul_idx) + 0.5, "label": "Peak: Jul ≈ 25°C"}])
# Plot — month order is a continuous temporal axis, so the ridges use the
# Imprint sequential gradient (imprint_seq: brand green -> blue) rather than
# a categorical palette; this keeps January anchored at #009E73.
plot = (
ggplot(ridge_df, aes(x="x", ymin="ymin", ymax="ymax", fill="month_idx", group="group"))
# No ribbon outline: the ymin edge would stroke a flat line across each
# ridge's full visible x-span, cutting through neighboring ridges. Instead
# only the top density curve (ymax) is stroked, drawn as a separate line.
+ geom_ribbon(alpha=0.85, color=None)
+ geom_line(aes(y="ymax"), color=INK_SOFT, size=0.5)
+ scale_fill_gradient(low="#009E73", high="#4467A3")
# geom_text from a separate dataframe anchored to July's y-band (showcases multi-layer grammar)
+ geom_text(
data=peak_df,
mapping=aes(x="x", y="y", label="label"),
inherit_aes=False,
color=ACCENT,
size=3.5,
fontweight="bold",
ha="right",
va="center",
)
# Diagonal leader segment from label anchor to July's density peak at (25, jul_idx+ridge_scale)
+ annotate(
"segment", x=25.5, xend=33.5, y=jul_idx + ridge_scale - 0.3, yend=float(jul_idx) + 0.5, color=ACCENT, size=0.8
)
+ scale_y_continuous(breaks=list(range(12)), labels=months, limits=(-0.5, 13.8))
+ labs(
x="Temperature (°C)",
y="Month",
title="ridgeline-basic · python · plotnine · anyplot.ai",
subtitle="Monthly temperature distributions — Northern Hemisphere temperate climate",
)
+ theme_minimal()
+ theme(
figure_size=(8, 4.5),
plot_background=element_rect(fill=PAGE_BG, color=PAGE_BG),
panel_background=element_rect(fill=PAGE_BG),
panel_border=element_blank(),
text=element_text(size=7, color=INK_SOFT),
axis_title=element_text(size=10, color=INK),
axis_text=element_text(size=8, color=INK_SOFT),
plot_title=element_text(size=12, color=INK, fontweight="bold"),
plot_subtitle=element_text(size=8, color=INK_SOFT),
plot_margin=0.03,
panel_grid_major_y=element_blank(),
panel_grid_minor=element_blank(),
panel_grid_major_x=element_line(color=INK, size=0.3, alpha=0.10),
legend_position="none",
)
)
# Save
plot.save(f"plot-{THEME}.png", dpi=400, width=8, height=4.5, units="in", verbose=False)
Part of Basic Ridgeline Plot on anyplot.ai.